SCRS Conference Proceedings on Intelligent Systems
Bone Fracture Detection in X-ray Images using Convolutional Neural Network
Authors: Rinisha Bagaria, Sulochana Wadhwani and A. K. Wadhwani
Publishing Date: 26-04-2022
ISBN: 978-93-91842-08-6
Abstract
It is critical to design a fracture detection system to offer quick results and reduce diagnosing errors. Using X-ray images in growing artificial intelligence methodologies, especially the deep learning method, has become a practical choice for detecting bone fractures. This research paper suggests a deep learning method using X-ray images for early diagnosis of bone disorders and also detection of different bone fractures. The effectiveness of the convolutional neural network model for classifying bone fractures from normal bones is used. Several significant factors such as no. of epochs, batch size, type of optimizers and learning rate are considered to find the best-suited model. Hence, it is found that the convolutional neural network model has good performance using the specificity of 89.865%, accuracy of 90% approximately, and area under ROC curve of 0.8088.
Keywords
Deep Learning, Medical Image Analysis, X-ray Imaging, Image Processing Techniques, Image Classification.
Cite as
Rinisha Bagaria, Sulochana Wadhwani and A. K. Wadhwani, "Bone Fracture Detection in X-ray Images using Convolutional Neural Network", In: Raju Pal and Praveen Kumar Shukla (eds), SCRS Conference Proceedings on Intelligent Systems, SCRS, India, 2022, pp. 459-466. https://doi.org/10.52458/978-93-91842-08-6-43